(old)Puppy Linux Discussion Forum

Posted: **Thu 08 Nov 2012, 22:17**

file:///usr/share/doc/HOWTO-regexps.htm wrote:[Note: Puppy has a regular expression evaluation and learning tool in the Utilities menu]
Once upon a time (version 1.04) I have used it much.
But now it seems to have disappeared

If you are running Python you can use redemo.py

I created redemo.sh and made it executable

Code: Select all

python /usr/lib/python2.6/site-packages/scripts/redemo.py

Posted: **Fri 09 Nov 2012, 00:51**

Full Install of 5.4.0.90 PAE to ext4 partition

Confirming that black screen no longer appears on bootup

Have installed about dozen pets from storage partition and everything works.

Posted: **Fri 09 Nov 2012, 01:22**

Somehow (new to forum) I think this has also gone somewhere else as well - appologies

Notifying BUG here as it concerns my test of 5.4.1 beta

Have just downloaded beta, tried on my Dell Vostro V360 all in one desktop. It boots to a totally black screen (nothing visible) which blinks the desktop on for about 0.5 seconds every 10-15 seconds. I also understand it affects some laptops as well of both Dell and other brands that use the nvidia optimus gt525 video chipset

This is a known bug with ubuntu 12.04 precise and I believe affects all kernels in all distros kernel 2.6 thru 3.5 that use a video driver unless it uses vesa. I understand the bug has been fixed in later 3.5 and up kernels

see this: https://bugs.launchpad.net/ubuntu/+sour ... bug/918769

and this which is too technical for me !! : https://bugs.freedesktop.org/attachment.cgi?id=62301

Has anyone else had this problem?

Hope this helps make Puppy better and more versatile

Posted: **Fri 09 Nov 2012, 07:05**

Peter.M wrote:Somehow (new to forum) I think this has also gone somewhere else as well - appologies

Notifying BUG here as it concerns my test of 5.4.1 beta

Have just downloaded beta, tried on my Dell Vostro V360 all in one desktop. It boots to a totally black screen (nothing visible) which blinks the desktop on for about 0.5 seconds every 10-15 seconds. I also understand it affects some laptops as well of both Dell and other brands that use the nvidia optimus gt525 video chipset

This is a known bug with ubuntu 12.04 precise and I believe affects all kernels in all distros kernel 2.6 thru 3.5 that use a video driver unless it uses vesa. I understand the bug has been fixed in later 3.5 and up kernels

see this: https://bugs.launchpad.net/ubuntu/+sour ... bug/918769

and this which is too technical for me !! : https://bugs.freedesktop.org/attachment.cgi?id=62301

Has anyone else had this problem?

Hope this helps make Puppy better and more versatile

I haven't tried beta .1 yet, but reported very similar video issues on page 16 of this thread, which apparently were not deserving of Barry's response.

If I hadn't had two monitors available, I wouldn't have been able to do any experimentation at all with Puppy 5.4. As it is, configuring a display was gruesomely difficult, and at times both monitors were blacked out, and only one was even marginally usable at the available settings.

And these are problems I've never encountered with any Puppy from 1.1 through 5.2.8.

Posted: **Fri 09 Nov 2012, 10:00**

otropogo wrote:I haven't tried beta .1 yet, but reported very similar video issues on page 16 of this thread, which apparently were not deserving of Barry's response.

Not the kind of comment that is going to get me "onside".

I often don't respond to a bug report simply because I haven't got a clue how to fix it. If someone reports a "black screen", there is not much to go on.

Often the only apparent solution is to use some other puplet or even other distro that does work with that hardware.

Posted: **Fri 09 Nov 2012, 10:06**

menu entries still not working 100% http://bkhome.org/blog2/?viewDetailed=00021

Posted: **Fri 09 Nov 2012, 10:09**

Precise Puppy 5.4.1 coming soon.

That's enough, I reckon that we have got this pup in good shape. I am going to do a build of 5.4.1-final right now, then do some sanity testing, then upload it.

The sanity-testing is required, as it is always possible that the recent fixes, such as the menu-category fix when installing packages, and the glib/gio module-loading fix, may themselves introduce some other bug -- even though they may seem like straight-forward fixes, I know from experience to be very cautious.

Well, it is even possible that my sanity test might miss some new subtle bug, but it is a question of the odds being in favour of everything being ok.

After 5.4.1 is released, the "Puppy community" is invited to take over and work toward 5.4.2+. If there is a consensus of developers, for a group or particular individual to coordinate, then go for it!

Posted: **Fri 09 Nov 2012, 10:32**

Um, here is an example of a potential last-minute bug...

I have put in the glib PET from akash_rawal, but I just looked at his compile configure:

./configure --prefix=/usr --sysconfdir=/etc --enable-debug=yes

...there is no "--build" parameter, so although the name of the PET specified that it is for "i686", there is no guarantee that this is so. The compile could be optimised for whatever CPU that akash_rawal has.

Therefore, I won't take the risk. I will build 5.4.1-final without akash_rawal's glib PET.

Posted: **Fri 09 Nov 2012, 14:51**

BarryK wrote: ./configure --prefix=/usr --sysconfdir=/etc --enable-debug=yes

Note "--enable-debug=yes" increases a size of compiled binary.
The default option is "--enable-debug=minimal" but if we want to get a smallest binary,
we must use "--enable-debug=no".
This also true for GTK and Pango libraries.

Posted: **Fri 09 Nov 2012, 17:05**

BarryK wrote:
otropogo wrote:I haven't tried beta .1 yet, but reported very similar video issues on page 16 of this thread, which apparently were not deserving of Barry's response.
Not the kind of comment that is going to get me "onside".

I often don't respond to a bug report simply because I haven't got a clue how to fix it. If someone reports a "black screen", there is not much to go on.

Often the only apparent solution is to use some other puplet or even other distro that does work with that hardware.

My report on page 16 described the most serious display configuration issues I've encountered in all the Puppies I've used since 1.1 - much more than a mere "black screen" report.

Ignoring bug reports because you dislike the poster simply undermines confidence in your work, as well as discouraging essential feedback.

Posted: **Fri 09 Nov 2012, 18:42**

01micko wrote: akash_rawal

To save you the hunt I've attached the glib patch I use in Slacko-beta series based on slackware-14.0. This is based directly on jemimah's work.

Cheers

Thanks.

I applied the patch and rebuilt glib, and abiword's help works

One slight problem, with an additional volume monitor running a mounted volume is listed twice in the filechooser sidebar. I modified it so that the native mount monitor is turned off on exporting "GIO_IGNORE_NATIVE_VOLUME_MONITOR" environment variable.

Package:
http://dl.dropbox.com/u/58347439/protec ... e-i686.pet

New patch:
http://dl.dropbox.com/u/58347439/protec ... 9.patch.gz

Posted: **Fri 09 Nov 2012, 18:46**

otropogo wrote:Ignoring bug reports because you dislike the poster simply undermines confidence in your work, as well as discouraging essential feedback.

Oh please!!!!

What nonsense.

This is one example of many that have contributed to my desire to withdraw from the "frontline" development.

Posted: **Fri 09 Nov 2012, 18:48**

Precise 5.4.1 has been released, and I have started a new Forum thread:

http://murga-linux.com/puppy/viewtopic.php?t=82092

Posted: **Fri 09 Nov 2012, 19:30**

BarryK wrote:
otropogo wrote:Ignoring bug reports because you dislike the poster simply undermines confidence in your work, as well as discouraging essential feedback.
Oh please!!!!

What nonsense.

This is one example of many that have contributed to my desire to withdraw from the "frontline" development.

Barry,
Please keep in mind that for every one of the people out there who acts like this... you have hundreds (if not thousands) who appreciate all the work you do. This entire community woudlnt exist if it werent for you. Please dont let the bad apples discourage you.

Posted: **Fri 09 Nov 2012, 19:55**

Q5sys wrote:
BarryK wrote:
otropogo wrote:Ignoring bug reports because you dislike the poster simply undermines confidence in your work, as well as discouraging essential feedback.
Oh please!!!!

What nonsense.

This is one example of many that have contributed to my desire to withdraw from the "frontline" development.
Barry,
Please keep in mind that for every one of the people out there who acts like this... you have hundreds (if not thousands) who appreciate all the work you do. This entire community woudlnt exist if it werent for you. Please dont let the bad apples discourage you.

I would differently agree on this one two, keep it up Barry and don't let anyone discourage you on what you have done which is to bring us a little puppy to play with.

Cheers

Posted: **Fri 09 Nov 2012, 20:26**

Trying more puppies recently I still cannot get my Huawei E160x 3G Modem
With Amaysim/Optus (Sydney,Australia) to work natively with puppy

All I need is the APN the phone number #99whatever and CHAP. that is fine

I have to revert to Mobile Partner version 11 or something.

In Precise it recognizes the modem and tells me so but even with the modem switch I cannot get, GSM connect to work for me. no ttyUSB,

Once Mobile Partner is installed, the USB_Mode_Switch Flipflopping the CD/Modem actually causes the modem to repeatedly be plugged in and then unplugged by the software a few times before Mobile Partner settles down and recognizes the modem, once connected Blinky works and I can view the info on statistics and see the Modem hardware next to etho and Lo.

I don't have any better luck with earlier puppies, Gnome network manager works fine in large distros, but I notices that Mobile Partner manages the toggle between mobile bands UMTS & HSDPA better that the Network manager

In the help file Barry mentioned that he used an E160 and wanted some more info so I add it here that it could help.

Now I'll get the new ISO.

Posted: **Fri 09 Nov 2012, 21:03**

Everything working nicely.

Not sure if this is a bug, but the usual compile line for Gtk programs didn't work.

Code: Select all

 gcc `pkg-config gtk+-2.0 --cflags` menu.c -o menu -lgtk-x11-2.0

Should the invocation be different?

EDIT: To answer my own question - Yes.

Code: Select all

gcc -o menu menu.c `pkg-config --libs --cflags gtk+-2.0`

Works

Cheers,
s

Posted: **Sat 10 Nov 2012, 03:00**

RE:

"I often don't respond to a bug report simply because I haven't got a clue how to fix it. If someone reports a "black screen", there is not much to go on.

Often the only apparent solution is to use some other puplet or even other distro that does work with that hardware."

Sorry Barry, I thought I was explicit enough re the problem of the black screen, and I am persevering with Puppy because I believe your majestic efforts have resulted in a superb compact distro.

As I said this problem affects most current distros and relates to the Nvidia "Optimus" technology for power savings on the GPU in dual display situations, usually with an integrated intel video and nvidia combination. I imagine this technology will rear it's head more and more and present ongoing problems for linux. The bumblebee project http://bumblebee-project.org/ is set up especially to solve this problem, and I have heard that a fix for the optimus problem is in place in the latest kernel version.

On Ubuntu Precise it is possible to boot into vesa mode by adding nomodeset on the command line on boot, even if the resolution is not ideal at least you can get a desktop displayed. I tried pressing F3 on Puppy live distro 5.4 beta boot and entering "puppy nomodeset pfix=vesa,ram" which should have had the same result, but to no avail the kernel did not appear to accept the command or at least did not put X into vesa mode.

If I can get nomodeset to work, then I can play around with a fix for the native video driver I assume that is nouveau being based on precise

Can anyone help with getting Puppy to boot to Vesa - I also tried vga=ask that changed the mode OK, text for boot came up in suitably small display but still went into graphics mode with the native driver active and black screen with blink.

So any suggestions anyone?

Posted: **Sat 10 Nov 2012, 03:28**

Hello Peter.M...and...Sulphurcocky.

Nice to see a few more Aussies on the forum.

Sulphurcocky....wireless etc.
tempestuous is a good person to ask.

rcrsn51...rerwin...and plenty of others provide solid advice.

Tethering...

http://www.murga-linux.com/puppy/viewtopic.php?t=77767

Chris.

Posted: **Sat 10 Nov 2012, 04:30**

I have uploaded an improved-but-not-yet-perfect version of the integrated Frisbee network manager, to make the pop-up messages more readable without slowing the initialization of networks. I also added the switching between versions of the network tray.

Peebee pointed out to me that the messages each delay network initialization by at least 6 seconds. And, because they all appeared at the same position on the screen, some could be obscured by others. I corrected that by placing the "announcement" messages at the bottom-left of the screen and the status messages at the bottom center. Messages will be delayed in those spots if another is being displayed. So, the messages may appear in sequence even though the processing is not delayed by them.

Previously, the network tray was selected only at boot time and only if Frisbee was the selected default "network setup tool". Now, when Frisbee is invoked through the "connect wizard", it will change to the Frisbee-network tray upon exit via the "exit" button. Canceling Frisbee by way of the "X" button does not trigger the swap. To switch back, the "woof-updates" package switches back to the normal network-tray icon if anything other than Frisbee is chosen.

However, there remains a malfunction that occurs the first time Frisbee is selected in a session. Upon exit, Frisbee appears to abort without running to the end of the script, so the swap does not occur. Subsequent selections of Frisbee seem to perform the swap. This behavior appears to cause no harm and corrects itself on an immediate second selection of Frisbee. I continue to investigate that.

The Frisbee and woof-update 20121108 packages are on page 15 here:
http://www.murga-linux.com/puppy/viewto ... 521#662521 for Frisbee
http://www.murga-linux.com/puppy/viewto ... 518#662518 for woof_updates

For testing, be sure to have the remaining packages at the woof_updates link also installed (except for the old woof_update-20121101).
Richard

UPDATE 11/10/2012: Due to faulty testing, the greyed-out wireless tab was erroneously announced as fixed. It is not fixed. The problem was that the indications of acquired interfaces were not being set. Please replace it with the now-corrected package at the above Frisbee link. Sorry for wasting your time.

UPDATE 11/11/2012: Uploaded new versions of the packages at the above links, to improve the usability. Please try them. Thanks.

UPDATE 11/12/2012: Uploaded new versions of the woof-updates and frisbee packages at the above links, to continue improvements to the usability. Modified Frisbee to allow it to run in slacko, as well.
R

(old)Puppy Linux Discussion Forum

Precise Puppy 5.4 and 5.4.2

Re: regular expression evaluation and learning tool

3G Modem

devx

Another Frisbee-3.0 update, 20121108.